[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-rt
Subject:    RE: RT-Linux update and questions
From:       "Andris Pavenis" <pavenis () laima ! acad ! latnet ! lv>
Date:       1997-09-30 17:22:52
[Download RAW message or body]

> 
> - As you called for "other problems" too: William Montgomery recently
> pointed out an IDE error:
> 
> >   Sep  8 17:19:34 samiam kernel: hda: write_intr: status=0xd0 { Busy }
> >   Sep  8 17:19:34 samiam kernel: ide0: reset: success
> 
> when running the fp example. I have the same, occasionally, both with
> your kernel and with Mantegazza's. Moreover I note that, for tasks using
> FPU, the jitter increases dramatically (ten times as much or more) with
> respect to plain integer/long long tasks.
> This issue is very interesting for me, as I am working on an axis
> control application which would be glad to use doubles with low jitter.
> 
I got similar message when I errorously used  cli() and sti()  in RT 
task (more exactly while processing the interrupt I hooked with  
request_RTirq() )   instead of use of r_cli() and r_sti(). I haven't tried 
this with FPU

Andris Pavenis 

[prev in list] [next in list] [prev in thread] [next in thread]